Comparisons of Endo-Wing-assisted Colonoscopy Versus Standard Colonoscopy in Improving Adenoma Detection Rate in Symptomatic Patients Suspected of Colorectal Cancer

Overview

Colorectal cancer (CRC) is the most common cancer in Malaysia. Colonoscopy is the gold standard for bowel cancer screening and the diagnosis of bowel pathology. Adenoma detection rate (ADR) is an important intraprocedural quality indicator, where endoscopists with ADR below 20% had ratios for the development of cancer more than 10 times higher than endoscopists with ADR more than 20%. A low ADR can be due to inferior visualization, especially at locations proximal to colonic folds and flexures. Endo-Wing™ (Shangxian Minimal Invasive Inc, China) is a medical-grade soft silicone rubber device that is attached to the end of a colonoscope. It has 6 soft wing-like projections that are bendable and pliable, which gives a superior visualization directly from its action of flattening the colonic folds and maintaining the central view of the colonoscope during withdrawal. This can improve the endoscopist's ADR. The purpose of this study is to compare the adenoma detection rate between standard colonoscopy and Endo-Wing™ assisted colonoscopy.

This study is a single-centre, non-blinded, randomised control trial done at the Hospital Canselor Tuanku Muhriz (HCTM), UKM. All HCTM patients undergoing screening, diagnostic, surveillance and therapeutic colonoscopy at the HCTM endoscopic suite will be included. The exclusion criteria are patients with colonic strictures, active colitis, known polyposis syndrome, known severe diverticulosis, known right hemicolectomy, patients with colostomy/ileostomy and poor bowel preparation. The study will be performed by 3rd and 4th-year surgical registrars who have completed their compulsory endoscopic training. The surgical review board and medical ethics committee of the Faculty of Medicine, Universiti Kebangsaan Malaysia (UKM) have reviewed and given their approval for this study to be conducted. All patients in this study have signed an informed and written consent. Patients who arrive at the HCTM endoscopy suite will be seen by medical officers and assessed for eligibility. Patients who are eligible and consented to study participation will be randomly allocated using block randomization into either of 2 groups: a standard colonoscopy or Endo-Wing colonoscopy. A standard protocol of 3 litres polyethylene glycol electrolyte solution will be used as bowel preparation at intervals. Conscious sedation will be administered according to standard protocol (IV Midazolam 3mg and IV Pethidine 25mg), and if the endoscopists deem the patient is in discomfort or pain, then an added dose of IV Midazolam 2mg and IV Pethidine 25mg will be given. Bowel preparation quality will be assessed using the Boston Bowel Preparation Scale (BBPS) during colonoscopy. All colonoscopies will be performed using Olympus Evis Exera III 90 series or Olympus Evis Lucera 260 series colonoscopes. If a patient is randomized to the Endo-Wing™ group, Endo-Wing™ will be attached to the distal tip of the colonoscope. In both groups, the advancement of colonoscopy will be done according to the standard insertion technique. Surgical registrars will be supervised throughout the procedures by a specialist surgeon. Once the caecum is intubated, the minimum time for withdrawal and examination of colonic mucosa is 6 minutes. Any adenoma/polyps that are seen will be recorded in terms of their presence, location, size and number. Location will classified according to proximal (caecum till transverse colon) and distal (descending colon till rectum) location. The size of the polyps will be divided into categories of 1-5mm, 6-9mm and more than 10mm. On the other hand, the number of adenomas seen will be divided into 1-2, 3-5 and more than 5 adenomas categories. Patients will be subsequently reviewed in the HCTM endoscopy suite post-colonoscopy for any immediate adverse effects. Once deemed well, patients will be discharged home and followed up with a telephone call 24 hours after the colonoscopic procedure and subsequently one week later. Randomization will be carried out using an R package (randomizeR) based on the central block randomization (1:1 ratio) of varying block sizes. The sequence of group assignments for each study participant will not be revealed to the participant recruiter to maintain allocation concealment. For the endoscopists, they will not know the procedures assigned to the participants until the commencement of the colonoscopy. The primary outcome of this study is the adenoma detection rate (ADR), which is defined as the proportion of colonoscopies with at least one adenoma detected to the total colonoscopies in that group. Secondary outcomes are the distribution of adenomas by location, number of adenomas detected, size of adenomas, caecal intubation rate, the total amount of sedation used and adverse effects. For sample size calculation, we plan to study the ADR (a proportion), with a one control(s) per experimental subject ratio. The estimated sample size will be calculated using a chi-square test procedure with an independent design based on the power and sample size (PS) software version 3.1.4.2. The study power and type I error rate are fixed at 80% and 0.05 (two-sided), respectively. Based on a recent study using a colonoscope adjunct to improve visualization with an ADR improvement from 28.8% (p0) to 49.1% (p1), with a true difference of 20.3. The calculated sample size is, therefore, 89 patients per arm with a total number of 178 patients. Taking into account a 10-percent drop-out rate, the total number of subjects to be recruited is 99 subjects per group (ntotal = 198 subjects) For statistical analysis, all the variables will be descriptively summarised using means and standard deviations for continuous variables and frequency and percentage for categorical variables. Comparison of categorical variables, adenoma detection rate and adenoma characteristics will be made using Pearson chi-squared and Fisher's exact tests. A P-value of 0.05 (two-sided) will be used as a significance threshold. Statistical analysis will be performed using IBM SPSS Version 25 software